全球变化知识资源中心

Ecological issues of China's Silk Road Economic Belt are receiving increasing attention. It is necessary to monitor vegetation changes in the Belt and explore their relationship with climate changes and human activities to ensure sustainable regional socioeconomic development. However, there is no adequate understanding of the relative importance of climate changes and human activities on vegetation over multiple timescales. In this study, ensemble empirical mode decomposition (EEMD) and residual trends (RESTREND) methods were used to establish the relative importance over multiple timescales. (1) The dominant timescales of vegetation changes have obvious spatial heterogeneity, which was affected by vegetation types and elevation. Grassland in high altitude areas, needleleaf forest, and shrubland were characterized by 3-year oscillations. Low-altitude areas mainly covered by cropland were characterized by long-term trends. (2) Climate types mainly affected the relationship between vegetation and climate changes at short timescales, while vegetation types, elevation, landforms affected the relationship over each timescale. Over the 3-year timescale, temperature and precipitation had a negative and a positive relationship respectively with vegetation changes in semi-arid and arid regions, but a positive relationship and a negative relationship respectively in semi-humid and humid regions. As the timescale widened, temperature had a positive relationship in all areas, except for needleleaf forest, and precipitation had a positive relationship except in high-latitude areas dominated by alpine meadow and in karst areas. (3) Vegetation changes were dominated by human activities at 3- and 6-year timescales but by climate changes in the long-term trend. However, when considering the total effects of these three timescales, climate changes dominated the overall changes of vegetation, while human activities also played an important role in vegetation reduction and insignificant changes. Vegetation changes were influenced by cross-scale climate changes. This study showed that multiple timescale analysis could deepen our understanding of the roles of climate and human activities on vegetation changes.